CeO2 Structure Adjustment by H2O via the Microwave–Ultrasonic Method and Its Application in Imine Catalysis

CeO2 with fusiform structures were prepared by the combined microwave–ultrasonic method, and their morphologies and surface structure were changed by simply adding different amounts of H2O (1–5 ml) to the precursor system. The addition of H2O changed the PVP micelle structure and the surface state, resulting in CeO2 with a different specific surface area (64–111 m2 g−1) and Ce3+ defects (16.5%–28.1%). The sample with 2 ml H2O exhibited a high surface area (111.3 m2∙g−1) and relatively more surface defects (Ce3+%: 28.1%), resulting in excellent catalytic activity (4.34 mmol g−1 h−1).


INTRODUCTION
Imine compounds are important intermediates for many biological, agricultural, and pharmaceutical compounds, such as alkaloids, membered heterocycles, and nitrogen heterocycles (Martin, 2009;Kaldas et al., 2017); thus, it is of great importance to develop a new synthesis approach for imine. CeO 2 is a new promising catalyst for imine synthesis, which can overcome the problems caused by traditional acid/base catalysts (Tamura and Tomishige, 2015) and bring a lot of advantages, such as mild synthesis conditions and fast separation. Many research studies have focused on CeO 2 application in imine synthesis (Geng et al., 2016;Long et al., 2019a;Long et al., 2019b;Cao et al., 2020;Rizzuti et al., 2020;Tamura and Tomishige, 2020), and it has been reported that CeO 2 morphologies can greatly influence the catalytic activity for imine synthesis (Zhang et al., 2017;Yang et al., 2018;Zhang et al., 2018;Yang et al., 2020a;Yang et al., 2020b), which have a close relationship with the CeO 2 specific surface area and Ce 3+ defects.
The preparation approach can effectively change the structure of CeO 2 , especially using the microwave method. Microwaves can heat the target solution by interacting with the solution molecules; thus, the solution can be heated fast and uniformly, which can effectively improve the product quality (Reddy et al., 2012;Li et al., 2021;Ma et al., 2021). Besides, other energies, such as UV-light and ultrasonic force, can work together with microwaves to achieve a new CeO 2 material structure via the Multifunctional Microwave Synthesis and Extraction Workstation (Yang et al., 2020a). Ultrasonic force is a kind of energy-accumulated mechanical vibration waves with thermal effect, mechanical effect, and cavitation effect (Thompson and Doraiswamy, 2000;Dalas, 2001;Shen, 2009), which can be used to accelerate the reaction rate and improve particle dispersion during material synthesis.
Microwave-or ultrasonic-assisted technology for CeO 2 synthesis has developed very rapidly in recent years (Leonelli and Mason, 2010;Phuruangrat et al., 2017;Zhao et al., 2018;Mousavi-Kamazani and Ashrafi, 2020;Chen et al., 2021;Zhai et al., 2021), and it has been found that different CeO 2 structures were obtained by using different energy inputs with the same solution (Yang et al., 2020a). Applying microwaves and ultrasonic waves together to the same solution can synthesize materials with a novel structure. In this way, the heat and mass can be transferred much better, and a new material structure can be achieved by changing the solution very slightly.
In this work, a series of CeO 2 nanomaterials was synthesized by a combined microwave-ultrasonic method. Different CeO 2 structures can be achieved only by changing the amount of deionized water in the solvent. The obtained CeO 2 exhibited different catalytic activities for imine synthesis, and their structures were further characterized by XRD, IR, SEM, XPS, and BET to figure out the materials' structure and the reason behind it.

Synthesis of CeO 2
All chemicals were provided by Adamas Reagents and used as received. The combined microwave-ultrasonic method was used to synthesis CeO 2 , using Ce(NO 3 ) 3 ·6H 2 O as the starting precursor, PVP as the structure directing agent, and acetic acid (HAC) as the mineralizer. The solvents were formed by ethylene glycol (EG) and DI-H 2 O, with a fixed total volume of 35.0 ml. The synthesis procedures were all the same; only the solvent content was changed by using different amounts of H 2 O, as shown in Table 1. The typical synthesis procedure was described using CeO 2 -H2 as an example. Ce(NO 3 ) 3 ·6H 2 O (2.17 g) was dissolved into the mixed solvent (33 ml EG + 2 ml DI-H 2 O) in a three-neck flask; then PVP (1.00 g) was added, and the resultant solution was stirred for 2 h. Finally, HAC (2.0 ml) was dropwise added, and the solution was further stirred for 15 min. Then the three-neck flask was transferred to a Multifunctional Microwave Synthesis and Extraction Workstation [Uwave-2000 from SINEO Microwave Chemistry Technology (China) Co. Ltd.] using microwave and ultrasonic force as the energy input, as shown in Figure 1A. The heating program is shown in Figure 1B, where the temperature is maintained at 100°C for 5 min and at 180°C for 12 min. After cooling naturally, the solids were centrifuged, washed by ethanol and DI-H 2 O four times, and dried in an oven (80°C) overnight. After that, they were calcined at 500°C (2°C/min) for 1 h in air to remove surface organic residues. In the sample name CeO 2 -Hx, x is the volume of DI-H 2 O added to the solvent.

Structure Characterization
The XRD patterns were acquired using a Shimadzu (Japan) D/Max-2500 diffractometer (nickel-filtered CuKα radiation). The morphologies of the CeO 2 samples were observed using a scanning electron microscope (SEM, Hitachi S-8000, Japan) in the secondary electron scattering mode at 5 kV. N 2 adsorption/ desorption was measured at 77 K by using a Micromeritics ASAP 2020 instrument, and all samples were degassed at 120°C for at least 5 h before testing. X-ray photoelectron spectra (XPS) were recorded using a Thermo Fisher ESCALAB 250 xi (England) using AlKα radiation (1,486.6 eV).

CeO 2 as Catalyst for Imine Synthesis
Calcined CeO 2 (50 mg) was added to a mixture of benzyl alcohol (10 mmoL) and aniline (20 mmoL). The resulting mixtures were stirred vigorously at 500 rpm at 50°C. After a certain time (at least 2 h), the imine product was analyzed using HPLC (Shimadzu LC-20AT, Japan) equipped with a Hypersil ODS C18 column (5 μm in a size of 4.6 mm × 250 mm), and the mobile phase was methanol: water = 8:2. In order to compare with the other reactions, the reaction rates are given as the amount of imine formation per hour per catalyst mass (mmol h −1 g −1 ).

RESULTS AND DISCUSSIONS
The x-ray diffraction (XRD) patterns of all samples are shown in Figure 2A. Typical CeO 2 crystalline phases (JCPDF No: 34-0394) appeared for all samples, which were prepared by the combined microwave-ultrasonic method and calcined at 500°C in air for 1 h, similar to the one reported before ), but the crystalline particle sizes were different when calculated by Scherrer equations based on the strongest (111) peak at 28.5°. The crystalline particle size changed in the range of 6.6-10.8 nm, which has a close relationship with the DI-H 2 O amount in the solvent. The CeO 2 -H0 sample, prepared without extra DI-H 2 O, possessed the biggest crystalline particle size of 10.8 nm. By adding extra DI-H 2 O to the precursor solution, the crystalline particle size decreased first (1-3 ml DI-H 2 O) and then increased again (4-5 ml DI-H 2 O). The smallest crystalline particle (6.6 nm) appeared when the extra DI-H 2 O was 3 ml (CeO 2 -H3).
The groups on the particle surfaces were characterized by FT-IR ( Figure 2B). After calcination at 500°C for 1 h, most of the organic groups were removed, and only a small amount of C=O (1,300-1,650 cm −1 ) and C-O (1,060 cm −1 ) groups can be observed besides the OH group at c.a. 3,500 cm −1 , indicating a relatively clean surface. These groups may be caused by the adsorption of CO 2 and H 2 O from the air, which can be further confirmed by XPS results in Figure 5.
The DI-H 2 O amount in the solvent can influence the morphologies of CeO 2 particles, as shown in Figure 3. When there was no extra H 2 O in the solution, the CeO 2 -H0 sample ( Figure 3A) exhibited an octahedral shape (diameter: 280 nm, length: 460 nm, and aspect ratio: 1.6), formed by aggregation of nanoparticles. The CeO 2 -H1 sample ( Figure 3B), with 1 ml extra DI-H 2 O in the solution, also showed a roughly octahedron structure, much more like a shuttle shape (diameter: 180 nm, length: 380 nm, and aspect ratio: 2), and small particles became more obvious in the aggregate. CeO 2 -H2 ( Figure 3C) possessed a similar structure to CeO 2 -H1; only the diameter became smaller, which was about 140 nm. When increasing H 2 O to 3 ml, CeO 2 -H3 ( Figure 3D) showed a smaller aggregation   structure, with a diameter of 80 nm and length of 200 nm (aspect ratio: 2.5). By further increasing H 2 O to 4 ml, the size of the aggregate in CeO 2 -H4 ( Figure 3E) increased again, exhibiting a structure similar to that of samples CeO 2 -H1 and CeO 2 -H2. Nevertheless, the octahedron/shuttle shape collapsed for CeO 2 -H5 ( Figure 3F); most were small particles, and only few aggregates can be observed. CeO 2 shuttles were previously synthesized by the hydrothermal method under similar conditions (Guo et al., 2008), but the size (diameter: 800 nm) was much larger than that of the samples in this work. The N 2 adsorption-desorption isotherm is used to characterize the materials' surface area, as shown in Figure 4A, and the pore diameter distribution is shown in Figure 4B. According to the IUPAC classification (Sing et al., 1985;Yang et al., 2020c;Zhang et al., 2021), the hysteresis loop of  the N 2 adsorption isotherm belonged to type VI, indicating that the materials were mesoporous. The specific surface area of CeO 2 -H2 (S BET : 111.3 m 2 /g) was similar to that of CeO 2 -H1 (S BET : 111.4 m 2 /g), larger than the values of other samples, while the minimum S BET (64.4 m 2 /g) belonged to CeO 2 -H5. Besides the surface area, the H 2 O amount also influenced the pore size distribution, as shown in Figure 4B. When there was no additional H 2 O added (sample CeO 2 -H0), the pore size distribution was relatively narrow, and almost all pore sizes were below 10 nm. When H 2 O was added to the system, the pore size distribution became broader, especially for sample CeO 2 -H2. This confirmed that adding H 2 O can influence the samples' morphologies. This may be related to the change in the PVP micelle structure. The extra H 2 O addition to the synthesis system changed the arrangement of PVP, resulting in the agglomeration alternation in CeO 2 particles, and the shape of CeO 2 became irregular along with H 2 O addition, as shown in Figure 3 and illustrated in Figure 6. The agglomeration variations in CeO 2 particles presented a different surface area and pore size distribution.
XPS is used to characterize the CeO 2 surface states ( Figure 5). Ce 3+ defects are important and have a close relationship with the catalytic activity of CeO 2 . In the Ce3d region ( Figure 5A), Ce 3+ ratios were calculated using the previous method, by fitting the area of Ce 3+ and Ce 4+ peaks (Zhang et al., 2017;Yang et al., 2018;Yang et al., 2020a). The CeO 2 -H0 sample had a Ce 3+ ratio of 24.2%. The addition of H 2 O increased the Ce 3+ ratio first and then decreased the number of Ce 3+ defects. The highest Ce 3+ ratio (32.0%) belonged to the CeO 2 -H3 sample, with 3 ml H 2 O added to the synthesizing system. CeO 2 -H2 possessed a Ce 3+ ratio of 28.1%, which was the second top of all Ce 3+ ratios. Besides, the O1s region can be fitted into three peaks: OH*/CO 3 2− , O 2 2− , and O 2− (Chen et al., 2013). OH*/CO 3 2− and O 2 2− were normally caused by the organic residuals or the surface impurity caused by CO 2 and H 2 O adsorption, and they can block the active sites of CeO 2 (Ferreira et al., 2012;Jiang et al., 2015;Yang et al., 2018). Thus, the higher the O 2− ratio is, the cleaner the CeO 2 surface is, indicating more active sites can be exposed. From the O1s region in Figure 5B, it can be seen that the CeO 2 -H2 sample possessed the highest O 2− ratio (68%), suggesting that it has the cleanest surface, which is beneficial to the catalytic activity. The highest O 2− ratio of CeO 2 -H2 may be because a suitable amount of H 2 O promoted the cleavage of organic residuals, as illustrated in Figure 6. This may compensate the slightly lower Ce 3+ ratio of the CeO 2 -H2 sample, resulting in the highest catalytic activity for imine synthesis, as shown in Figure 7. From the above-mentioned characterization, the addition of H 2 O can change the morphology of CeO 2 , which can further alter the specific surface area and the surface Ce 3+ defects. These changes may be related to the PVP micelle structure change and the CeO 2 surface coordination state induced by H 2 O, as sketched in Figure 6. When no H 2 O is added to the system, the CeO 2 surface is covered with EG molecules, and the surface is hydrophobic. Thus, the PVP formed a micelle structure, with the hydrophobic ends toward the CeO 2 agglomeration particles. When H 2 O is added to the system, it can react with the Ceprecursor, forming a Ce-OH structure; thus, the CeO 2 surface transformed from hydrophobic to hydrophilic, and the PVP formed a micelle structure, with the hydrophilic ends toward to the CeO 2 particles. Besides, Ce-OH can also yield a cleaner surface, and the condensation between -OH can happen to yield oxygen vacancies and H 2 O, yielding a higher Ce 3+ ratio. Meanwhile, H 2 O can also be adsorbed onto the CeO 2 surface, which can interact with the microwave and sonic energy much better than EG can, causing more wrinkles on the CeO 2 particles to increase the specific surface area. However, when more than 4 ml of H 2 O was added into the system, the PVP micelle structure was destroyed, and no shuttle structure can be observed.
When synthesized CeO 2 was used for imine synthesis, they exhibit different activities (Figure 7). The order of imine  conversion is CeO 2 -H2 (4.34 mmol g −1 h −1 ) ≈ CeO 2 -H3 (4.33 mmol g −1 h −1 ) > CeO 2 -H1 (3.78 mmol g −1 h −1 ) > CeO 2 -H4 (2.48 mmol g −1 h −1 ) > CeO 2 -H5 (1.72 mmol g −1 h −1 ) > CeO 2 -H0 (1.47 mmol g −1 h −1 ). The reaction rates of all the samples were far better than the 0.46 mmol g −1 h −1 reported in Angew (Tamura and Tomishige, 2015). It has been reported earlier that CeO 2 can be used as an easily separable catalyst for imine synthesis at low temperature (<100°C) (Zhang et al., 2017;Zhang et al., 2018;Chen et al., 2020;Wu et al., 2021) and oxygen vacancies (Ce 3+ defects) played an important role during this reaction. The Ce 3+ vacancies can promote the transition of benzyl alcohol to benzaldehyde, and the latter can easily couple with aniline to form an imine compound (Tamura and Tomishige, 2015;Zhang et al., 2017;Qin et al., 2019). From the above-mentioned results, it can be seen that CeO 2 -H2 possessed the largest surface area (111.4 m 2 /g) and a relatively high Ce 3+ ratio (28.1%), which contributed to the highest imine reaction rate. CeO 2 -H3 exhibited a similar catalytic activity, which may be due to its highest Ce 3+ ratio (32.0%), but its surface area (72.3 m 2 /g) is a little bit lower than that of CeO 2 -H2. The results confirmed that both surface area and Ce 3+ ratio are important to the CeO 2 catalytic properties for imine synthesis.

CONCLUSION
Different sizes of fusiform CeO 2 were synthesized by the combined microwave-ultrasonic method, and H 2 O was added to change the structure of CeO 2 . By changing the amount of H 2 O from 1 to 5 ml, the specific surface area of CeO 2 changed from 111 m 2 /g to 64 m 2 /g, while the Ce 3+ ratio varied in the range of 16.5% and 32.0%. The sample with 2 ml H 2 O has the largest surface area (111.4 m 2 /g) and a relatively high Ce 3+ ratio (28.1%), indicating that the CeO 2 -H2 sample has the largest amount of exposed surface oxygen vacancies, and the imine reaction rate is 4.34 mmol g −1 h −1 , which is almost three times the rate (1.47 mmol g −1 h −1 ) produced by the sample without addition of water.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
XC and JY conceived this work, HD collected the data. XC wrote the original manuscript, and JY reviewed and edited the manuscript.

FUNDING
This research was supported by the National Natural Science Foundation of China (Grant No. 12175035).